Automobile part intelligent management and fault tracing method and system

Through the intelligent management and fault traceability system of automotive parts, multi-dimensional scoring models and machine learning algorithms, combined with blockchain and QR code technology, the problems of low maintenance efficiency and inaccurate fault traceability are solved, and the problem of inaccurate fault traceability is achieved. Fast and accurate fault location and maintenance, improving maintenance efficiency and vehicle safety.

CN119066555APending Publication Date: 2024-12-03BAOTOU KAIYUAN DIGITAL CO LTD
View PDF 0 Cites 6 Cited by

Patent Information

Application Number
CN202411182433.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Traditional automotive parts management and maintenance efficiency is low, there are information missing and experience misjudgment in fault detection and analysis, resulting in maintenance delays and increased costs, and the fault traceability method cannot quickly and accurately locate the root cause of the problem.

Method used

Design an intelligent management and fault traceability system for automotive parts, including data acquisition and storage module, component importance grading module, vehicle fault inspection and analysis module, information traceability module and real-time monitoring and feedback module, using multi-dimensional scoring model, Machine learning algorithms, blockchain and QR code technology realize automated management and rapid fault location.

Benefits of technology

Improve the efficiency of automotive parts maintenance, reduce maintenance costs, ensure the safety performance of the vehicle, quickly identify and handle important component failures through automated analysis and real-time monitoring, reduce labor costs and time, and enhance the reliability and safety of the vehicle. sex.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066555A_ABST
    Figure CN119066555A_ABST
Patent Text Reader

Abstract

The invention provides an automobile part intelligent management and fault tracing method and system, and the system comprises a data collection and storage module, a part importance grading module, a vehicle fault inspection and analysis module, an information tracing module, and a real-time monitoring and feedback module. The method comprises the steps that data in automobile production operation are collected and stored, a single-automobile two-dimensional code, a VIN number and an independent two-dimensional code of each part are formed, and the importance degree of each part is graded through a machine learning algorithm; when an automobile breaks down, automobile sensor information and a fault code DTC are collected for fault reason analysis, related parts are checked according to part importance grade sorting, a code scanning device is used for scanning codes to obtain related full life cycle information of the parts, and corresponding measures are taken after traceability. Emergency situations are prevented by monitoring important parts, problems can be quickly positioned, a processing scheme can be given, and the safety and stability of the vehicle are effectively improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the Internet of Things, blockchain, QR code and artificial intelligence technology, and is applied to the automotive industry and information technology fields. It is a method and system for intelligent management and fault tracing of automotive parts. Background Art

[0002] With the continuous development of the automotive industry, the types and complexity of vehicle parts are increasing, and vehicle maintenance and management has become a complex task. Traditional vehicle parts management and maintenance mainly rely on manual records and the experience of maintenance personnel, which is not only inefficient but also prone to errors, especially in fault detection and analysis, which often leads to maintenance delays due to missing information or misjudgment due to past experience. In addition, traditional fault tracing methods cannot quickly and accurately locate the root cause of the problem, resulting in extended maintenance time and increased costs.

[0003] Therefore, designing an intelligent system that can automatically manage auto parts, detect and analyze faults in real time, and realize information storage and traceability of auto parts throughout their life cycle has important application value, helps improve auto maintenance efficiency, reduce maintenance costs, and ensure vehicle safety performance. Summary of the invention

[0004] The object of the present invention is to provide a method and system for intelligent management and fault tracing of automobile parts to solve the problems raised in the above-mentioned background technology.

[0005] In one aspect, the present invention provides an intelligent management and fault tracing system for automobile parts, comprising the following modules:

[0006] Data acquisition and storage module, used to collect and store information of various parts in the automobile production process;

[0007] The component importance grading module uses a component importance grading algorithm to dynamically score and grade components based on multi-dimensional indicators, including component function, historical failures, maintenance costs, usage frequency, supply chain data, and user feedback. The grading is to divide components into different importance levels.

[0008] Vehicle fault inspection and analysis module, used to perform fault analysis based on vehicle sensor information and fault codes;

[0009] Information tracing module, used to troubleshoot component failures based on the failure mode history library and real-time diagnostic data, and generate maintenance recommendations;

[0010] The real-time monitoring and feedback module is used to monitor the status of components during vehicle operation in real time and generate alarms and report problems in production.

[0011] Furthermore, the data collection and storage module also collects the production operation plan and supporting detailed data of a single vehicle, integrates the single vehicle VIN code as the main body to form a single vehicle file, and stores the data in the system database and / or the cloud.

[0012] Furthermore, the data acquisition and storage module performs AES symmetrical encryption on the collected component information, enters it into the system database, uploads it to the data cloud for backup, generates a QR code, and embeds it into the surface of the component or on the packaging.

[0013] Furthermore, the data collection and storage module collects workstation information, worker work information and operation information, stores it on the blockchain platform, generates transactions and packages them in blocks. The blockchain type is a consortium chain, which restricts access rights on the chain and is maintained by a dedicated person in the department.

[0014] Furthermore, the component importance grading algorithm in the component importance grading module includes the following steps:

[0015] 1) Collect functional information, historical failure information, maintenance cost information, usage frequency information, supply chain data information, and user feedback information of parts;

[0016] 2) Establish a multi-dimensional scoring model to score each component based on its key functions, historical failure rate, usage frequency, supply chain complexity, maintenance cost, and user feedback;

[0017] 3) Apply a weighted algorithm to the scoring results to calculate the importance score, and determine the importance level of the parts according to the preset threshold, and divide them into A, B, C, and D levels in descending order of importance;

[0018] 4) Generate a hash value for the full life cycle information of grade A parts and store it on the blockchain network, and associate it with the system database.

[0019] Furthermore, the component importance grading module dynamically adjusts the scoring results and updates the component importance level when new data is generated or usage changes.

[0020] Furthermore, the component importance grading module dynamically adjusts the scoring system in combination with a machine learning algorithm, and continuously optimizes the weights of the scores of each dimension based on newly input data.

[0021] Furthermore, the machine learning algorithm adjusts the scoring weights through the time weighting factor and the sensitivity coefficient, so that the importance score changes dynamically with the input of new data.

[0022] Furthermore, the component importance grading module records the whole life cycle information of components with importance level A in the form of hash values ​​on the blockchain; dynamically monitors the components with level A, collects component operation data through sensors, establishes a prediction model, and regularly inspects important components based on the model budget results to generate reports;

[0023] Furthermore, the vehicle fault inspection and analysis module is used to collect vehicle sensor information and fault codes DTC, and analyze the cause of the fault to accurately locate the source of the fault.

[0024] Furthermore, the information tracing module uses an association rule mining algorithm to match the fault mode history library with real-time diagnostic data, conduct component fault troubleshooting, obtain component-related information through QR code scanning, generate corresponding maintenance suggestions, and automatically update the fault mode history library and diagnostic model.

[0025] Furthermore, the real-time monitoring and feedback module is used to monitor the status of key components during vehicle operation in real time and automatically generate an alarm when an abnormality is detected.

[0026] Furthermore, the real-time monitoring and feedback module supports operators, inspectors and on-site technicians to report quality issues in the system, and designates responsible units to handle the issues through the system task allocation function.

[0027] Furthermore, the above module includes the following functions and contents:

[0028] The data collection and storage module collects automobile production operation plans, supporting detailed data, and parts information in the module. Specifically, the parts information includes part number, part name, quantity, supplier information, part production batch information, identification information, work station information, operator information, online and assembly time information, and final inspection confirmation form. The parts information is encrypted using a secret key using the AES symmetric encryption algorithm and then entered into the system database and uploaded to the data cloud for backup. A QR code is generated and embedded in the surface or packaging of the part. The QR code is scanned using a dedicated scanning device or a smart phone to obtain the full life cycle information of the part.

[0029] Specifically, the module collects work station information, worker work information and operation information: the work station information includes the work station setting work condition specifications and the material list displayed by work station; the worker work information means that each worker must sign in by facial recognition at the designated work station before going to work. If the sign-in fails due to mismatch, facial recognition verification will be performed at irregular intervals during work to record and verify the time and identity information of the worker's arrival at the workstation; the operation information means that when workers start to produce parts, they need to electronically sign the process sheet and completion sheet of the parts to confirm the completion of each operation step of the parts. The above-mentioned entire process data will be uploaded to the blockchain system in real time, and the generated transactions will be packaged in the blocks. The blockchain type is a consortium chain, which restricts access rights on the chain and is maintained by dedicated personnel within the department.

[0030] Furthermore, the automobile production operation plan, supporting detailed data, and parts information are all linked to the vehicle VIN code to form a vehicle file. The VIN code is a unique identification code for each vehicle. All of the above information can be quickly obtained by querying the VIN code.

[0031] Furthermore, key data in the collected data is extracted to generate a vehicle data acquisition card. The acquisition card has a QR code embedded in the vehicle body. The QR code can be scanned to obtain the single vehicle production operation plan, supporting detailed data and the part number, supplier, batch and number data of all assembled parts. The information is encrypted and stored in the system database and data cloud. The corresponding data can be directly obtained by scanning the code with a barcode scanner or a workstation scanning device.

[0032] The component importance grading module determines the component importance level through a component importance grading algorithm. Specifically, the algorithm includes the following steps:

[0033] 1. Data collection: collect historical failure data, production information, usage data and other relevant parameters of parts, mainly including the following data:

[0034] Component function description: the main function of the component and its role in the car;

[0035] Historical failure rate: historical failure data of components in different models, including detailed information of each failure;

[0036] Repair cost: the cost of repairing or replacing parts, including labor cost and material cost;

[0037] Frequency of use: The frequency of use and workload of components, which can be obtained through sensor data and usage records;

[0038] Supply chain data: The complexity and stability of the parts supply chain, including the number of suppliers, geographic location, and supply chain levels;

[0039] User feedback: User satisfaction and complaint rates for parts can be obtained through questionnaires, social media, and customer service records.

[0040] 2. Establish a multi-dimensional scoring model to score each component from the following dimensions:

[0041] Key Functionality Score (KFS): Assess the criticality of a component's function in a car. The score is generated through analysis of past data. The KFS ranges from 0 to 1. The higher the criticality, the higher the score.

[0042] Historical Failure Rate Score (HFRS): Based on historical failure data, the failure rate of each component is calculated. The failure rate score ranges from 0 to 1 and is calculated using the following formula:

[0043]

[0044] Maintenance Cost Score (MCS): A score is assigned based on the cost of repairing or replacing parts. C represents the maintenance cost. The maintenance cost score ranges from 0 to 1 and is calculated using the following formula:

[0045] MCS=1-e -C ;

[0046] Usage Frequency Score (UFS): A score is assigned based on the frequency of use and workload of a component. The Usage Frequency Score ranges from 0 to 1 and is calculated using the following formula:

[0047]

[0048] Supply Chain Complexity Score (SCCS): Scored based on the complexity and stability of the supply chain, n represents the number of suppliers, L represents the supplier level, and the supply chain complexity score ranges from 0 to 1 and is calculated using the following formula:

[0049]

[0050] User Feedback Score (US): Score based on user feedback satisfaction and complaint rate. The user feedback score ranges from 0 to 1 and is calculated using the following formula:

[0051]

[0052] 3. Weight distribution: Invite 8 experts, including 2 experts in the fields of automotive engineering, parts manufacturing, automotive maintenance, and market analysis, to participate in determining the weight of each dimension. First, conduct the first round of questionnaires, send questionnaires to the expert group to determine the initial weight of each dimension; collect the questionnaires and summarize them, calculate the average of the recommended weights for each dimension, and feedback the calculation results to each expert; start the second round of questionnaires, ask the experts to re-assign weights to each dimension based on the results of the previous round, collect the questionnaires and count the results again, calculate the average and feedback the results to the experts; continue three rounds of questionnaires, and feedback the summary results between each round of surveys. Experts in each round can adjust their weight suggestions until the expert group reaches a certain consensus and the opinions tend to stabilize. At this time, the finalized weights of each dimension are output.

[0053] 4. Calculate the comprehensive score of each component based on the score and weight of each dimension.

[0054] 5. Based on the comprehensive score, the parts are divided into four grades: A, B, C, and D. Grade A: comprehensive score ≥ 0.8, Grade B: 0.8>comprehensive score ≥ 0.7, Grade C: 0.7>comprehensive score ≥ 0.5, Grade D: 0.5>comprehensive score.

[0055] Furthermore, a machine learning algorithm is used to dynamically adjust and optimize the multi-dimensional scoring model, and the weight distribution of each dimension score is optimized based on the continuous input of new data. The algorithm steps are as follows:

[0056] 1. Initially, the weights of the scoring parameters of each dimension have been determined as W 0 ={ω 0,1 ,ω 0,2 ,...ω 0,n}, where ω 0,i represents the initial weight of the i-th parameter and satisfies:

[0057]

[0058] Scoring parameters include: key functionality score, historical failure rate score, repair cost score, usage frequency score, supply chain complexity score, and user feedback score;

[0059] 2. Collect and input new real-time data generated by each component during use into the model, D t ={d t,1 ,d t,2 ,...,d t,n}, where d t,i Represents the new data of the i-th parameter at time t. The new data is input into the scoring formula of each dimension to calculate the score, and the score C of each dimension of the part is output. t ={c t,1 ,ct,2 ,...,c t,n};

[0060] 3. Set a change sensitivity coefficient S for each scoring parameter i , which indicates the sensitivity of this parameter to the importance of the component, and its value range is 0 to 1. The higher the coefficient, the greater the impact of this parameter on the change of importance; the time weighting factor T is introduced t , whose value decreases over time, to reduce the impact of historical data on current calculations:

[0061]

[0062] Where λ is the decay rate parameter, t 0 is the initial time point;

[0063] 4. At each time point t, the weight of each dimension score is adjusted according to the deviation between the newly obtained score data of each dimension and the historical score data.

[0064] The adjustment formula is:

[0065]

[0066] Where: η is the learning rate, which controls the adjustment pace, is the historical average of the ith parameter, is the historical standard deviation of the i-th parameter; the adjusted score weight ratio is:

[0067] ω t+1,i =ω t,i +Δω t,i ;

[0068] 5. Normalize to ensure that the adjusted weights of each score still satisfy the constraint that the sum is 1:

[0069]

[0070] 6. Output the results.

[0071] Through the above-mentioned machine learning algorithm, it can be achieved that when the score of any component in a certain dimension continues to increase, the weight of the score will also continue to increase. Improving both aspects simultaneously can quickly improve the importance level of the component, so that the components that the staff considers to be unimportant based on experience can be quickly taken seriously. When relevant personnel find this situation, they should verify the situation of the component in time, and conduct troubleshooting and repair when necessary, and keep paying attention to it in the future. This method can realize intelligent sorting of the importance of automotive parts, and adjust and update it in real time. Through automated analysis, maintenance personnel can give priority to troubleshooting those components with high importance levels when facing vehicle failures, thereby locating problems faster and more accurately, completely getting rid of the traditional method of relying on human experience and judgment, and avoiding misjudgment caused by lack of experience or subjective bias. This not only improves maintenance efficiency, but also significantly reduces the time and labor cost of troubleshooting, providing a stronger guarantee for the safety and stability of the vehicle.

[0072] Preferably, the importance of each component is determined using a component importance evaluation and grading algorithm, and the components are divided into A, B, C, and D levels in descending order of importance. The full life cycle information of components with level A is generated into a hash value and stored on the blockchain network for comparison and verification when the code is scanned to obtain information in the future to prevent the information from being tampered with. The blockchain type is a consortium chain, which is maintained by a dedicated person in the department. The information index and verification information on the consortium chain are associated with the component QR code and stored in the system database. After scanning the QR code with a scanning device, the component information can be retrieved through the consortium chain network. Use a machine learning algorithm to dynamically monitor components with level A, collect component operation data through sensors, establish a prediction model, regularly inspect important components based on the model budget results, and generate reports to prevent vehicle failures caused by sudden damage to components that affect normal use.

[0073] Preferably, the workstation information, worker operation information, and the full life cycle information of parts with importance level A are generated and packaged in blocks, and the blockchain type is a consortium chain; to reduce the data pressure on the chain, the workstation information and worker operation information are stored on one chain, and the full life cycle information of parts with importance level A is stored on another chain; each chain is maintained by a dedicated person from each department, and cross-chain data interaction is performed by having a dedicated person from each department act as a notary, so that the information on the chain can flow across chains.

[0074] The vehicle fault inspection and analysis module: When the electronic control system of the car detects an abnormal situation, it will generate a specific diagnostic fault code DTC (Diagnostic Trouble Code) to indicate the type and location of the fault, and analyze the cause of the fault by collecting the car sensor information and the fault code DTC. The module will first monitor and summarize the data from various sensors of the car in real time. Through detailed analysis of these data, the module can identify potential problems and abnormal situations; at the same time, the fault code DTC will be collected and interpreted to provide specific information about the fault. These fault codes serve as an important basis for diagnosis and help to accurately locate the source of the fault.

[0075] The information tracing module: uses real-time diagnostic data and historical fault data, and uses the Apriori algorithm for association rule mining. The Apriori algorithm is an algorithm for association rule mining. It scans the data multiple times to find frequent item sets, and then derives association rules to discover potential associations in the data. The Apriori algorithm is used to analyze the relationship between diagnostic codes and component failures. According to the importance level of the components and the specific fault conditions, high-importance and high-impact components are checked first; matching faults, using past fault information and common problems of the entire related vehicle model to establish a fault mode history library, matching current fault symptoms with historical fault modes, and quickly locating faulty components; using a scanning device to scan the code to obtain all relevant information of the component, including process design process information, processing process information, assembly process information, including transportation, installation and maintenance records, combined with the hash value on the blockchain to verify its complete life cycle information, and take corresponding processing measures.

[0076] The real-time monitoring and feedback module: For any process or quality problems, operators, inspectors, and on-site technicians have the authority to report them in the system. After the problem is reported, the task decomposition personnel will designate the responsible unit in the system to handle it. The responsible unit must provide feedback in the system after the handling is completed. The process can only be ended after the decomposition personnel confirms that the problem is closed. The system supports the decomposition of multiple responsible units for the same problem and data report query functions; during the operation of the vehicle, the status of key components is monitored in real time, and anomalies are discovered through sensor data. When the system detects an anomaly, it automatically generates an alarm to prompt the driver and maintenance personnel to check, and feedback the actual fault handling situation to the system, improve the fault mode history library and diagnostic model, and prepare for the next precise fault positioning.

[0077] In another aspect of the present invention, the present invention provides the following technical solution: a method for intelligent management and fault tracing of automobile parts, the method comprising the following steps:

[0078] Data collection and storage: collect the production operation plan, supporting detailed data, and parts information of each vehicle, and integrate them into single-vehicle file storage based on the single-vehicle VIN code; collect and store work station information, worker work information, and operation information;

[0079] Component importance grading: dynamically grading the importance of automotive components, regularly inspecting A-level components, and generating reports;

[0080] Vehicle fault inspection and analysis, collecting vehicle sensor information and fault codes, and analyzing fault causes;

[0081] Information tracing, analyzing the relationship between fault codes and component failures, scanning codes to obtain relevant information about components, and taking action;

[0082] Real-time monitoring and feedback: real-time monitoring of the status of components during vehicle operation and generation of alarms.

[0083] According to the above content, the present invention has the following advantages and beneficial effects:

[0084] 1. When collecting data, collect the key data of the production operation plan, supporting details and parts information of the single vehicle of the automobile. Each part is equipped with a QR code, which is embedded on the surface or packaging of the part. By scanning these QR codes with a dedicated scanning device or a smart phone, the full life cycle information of the parts can be obtained immediately. All information related to automobile production, including the operation plan, supporting details and parts information, are associated with the unique identification code of the single vehicle - the VIN code, forming a comprehensive single vehicle file. You only need to query the VIN code to quickly obtain all information related to the vehicle; in addition, the whole vehicle data acquisition card is also designed to have a QR code embedded in the vehicle body. Scanning this QR code can obtain the single vehicle production operation plan, supporting details and the part number, supplier, batch and number information of all assembled parts. Whether using a barcode scanner or a workstation barcode scanner, the scanning operation can directly obtain the corresponding data. In the process of information tracing, by querying the vehicle's VIN number or scanning the whole vehicle data acquisition card, you can quickly and comprehensively obtain all the information of the vehicle, which is simple and efficient; scanning the QR code information on the surface of the part can quickly obtain the full life cycle information of the part, which is simple and convenient to operate. This method not only improves the efficiency of vehicle information management, but also provides reliable information technology support for automobile production and maintenance.

[0085] 2. By using a multi-dimensional scoring model to evaluate key components and combining machine learning algorithms to dynamically adjust the scoring weights of each dimension, real-time adjustment and intelligent sorting of the importance level of automotive components are achieved. This method completely breaks away from the limitations of traditional reliance on human experience and judgment, and can quickly and automatically identify and give priority to important components. When problems occur with components that were considered unimportant based on experience in the past, the system can also use machine learning algorithms to quickly discover and provide feedback, thereby improving the efficiency and accuracy of troubleshooting and shortening vehicle maintenance time; when facing vehicle failures, the first thing to do is to check the high-importance components related to the problem, so as to locate the problem faster and more accurately, effectively reduce maintenance costs and time, significantly enhance the safety and stability of the vehicle, and provide a strong guarantee for the reliable operation of the vehicle.

[0086] 3. For components rated as Class A, a dynamic monitoring mechanism is adopted. Through sensors installed in the vehicle, the operating data of these components are continuously collected. The collected data is input into the prediction model and a report is generated. Based on the report data, the system can identify possible signs of failure in advance, so as to arrange regular component inspection and maintenance, thereby avoiding vehicle breakdown due to component failure during driving. This preventive maintenance improves the reliability and driving safety of the vehicle and reduces the time and cost required for emergency repairs.

[0087] 4. The full life cycle information of parts with importance level A is recorded in the form of hash values ​​on the alliance chain maintained by specific personnel in the department to prevent the information from being tampered with; each worker's workstation sign-in information before taking up the post, the process sheet when producing parts, and the electronic signature information on the completion sheet are uploaded to another alliance chain maintained by specific personnel in the department. This method ensures the traceability and immutability of information, and also provides convenient permission control for data management. The adoption of a dual-chain structure significantly reduces the pressure on data on the chain, and is conducive to the flexible management of data permissions. By having designated personnel in the department act as notaries, cross-chain data interaction is realized, breaking the data island within the chain. On the basis of the immutable and traceable characteristics of the blockchain, the flow of cross-chain data is realized, which fully improves the security and reliability of information management.

[0088] 5. When operators encounter problems at any stage of the operation process, they can submit feedback to the system at any time. The system will automatically decompose the problem and transmit it to the relevant departments for resolution. The whole process is simple and efficient. During the operation of the car, the sensor will collect data and detect anomalies in real time. When the system detects an anomaly, it will automatically generate an alarm to prompt the driver and maintenance personnel to check and feedback the actual fault handling situation to the system. This not only improves the fault mode history library and diagnostic model, but also makes full preparations for the next precise positioning of the fault, thereby greatly improving the safety of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] Figure 1 This is the overall structural block diagram of the automobile parts intelligent management and fault tracing system of the present invention.

[0090] Figure 2 The present invention is a flowchart of the steps of the method for intelligent management and fault tracing of automobile parts. DETAILED DESCRIPTION

[0091] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0092] The present invention provides Figure 1 An automobile parts intelligent management and fault tracing system shown includes a data acquisition and storage module, a parts importance classification module, a vehicle fault inspection and analysis module, an information tracing module, and a real-time monitoring and feedback module;

[0093] The data collection and storage module is used to collect automobile production operation plans, supporting detailed data, and parts information. Specifically, the parts information includes part number, part name, quantity, supplier information, part production batch information, identification information, work station information, operator information, online and assembly time information, and final inspection confirmation form. The part information is encrypted using the AES encryption algorithm and then entered into the system database and uploaded to the data cloud for backup. A QR code is generated and embedded in the surface or packaging of the part. The QR code can be scanned using a dedicated scanning device or a smart phone to obtain the full life cycle information of the part.

[0094] Specifically, the module collects work station information, worker work information and operation information: the work station sets work conditions and specifications, the bill of materials is displayed by work station, and each worker must sign in by facial recognition at the designated work station before starting work. If the sign-in fails due to a mismatch, facial recognition verification will be performed at irregular intervals during work to record and verify the time and identity information of the workers' arrival at the workstation. When workers start producing parts, they need to electronically sign the process sheet and completion sheet of the parts to confirm the completion of each operation step of the parts. The above entire process data will be uploaded to the blockchain system in real time, and the generated transactions will be packaged in blocks. The blockchain type is a consortium chain, which restricts access rights on the chain and is maintained by dedicated personnel within the department.

[0095] What needs to be specifically explained in this embodiment is that the production operation plan, supporting detailed data, and parts information of a single vehicle are all associated with the single vehicle VIN code to form a single vehicle file. The VIN code is the unique identification code of each single vehicle. All the above information can be quickly obtained by querying the VIN code. Furthermore, the key data in the collected data is extracted to generate a vehicle data acquisition card. The acquisition card has a QR code embedded in the vehicle body. The QR code is scanned to obtain the single vehicle production operation plan, supporting detailed data, and the part number, supplier, batch, and number data of all assembled parts. The information is encrypted and stored in the system database and data cloud. Through a barcode scanner or a workstation barcode scanner, the corresponding data can be directly obtained by scanning the code.

[0096] The component importance grading module is used to dynamically assess the component importance levels using the component importance grading algorithm.

[0097] Specifically, the component importance grading algorithm includes the following steps:

[0098] 1. Data collection, collect the following data: component function description, that is, the main function of the component and its role in the car; historical failure rate, that is, the historical failure data of the component in different models, including detailed information of each failure; maintenance cost, that is, the cost of repairing or replacing components, including labor cost and material cost; frequency of use, that is, the frequency of use and workload of components, which can be obtained through sensor data and usage records; supply chain data: the complexity and stability of the component supply chain, including the number of suppliers, geographical location, and supply chain level; user feedback, that is, user satisfaction and complaint rate with components, which can be obtained through questionnaires, social media, and customer service records.

[0099] 2. Establish a multi-dimensional scoring model to score each component from the following dimensions:

[0100] Key Functionality Score (KFS): Assess the criticality of a component's function in a car. The score is generated through analysis of past data. The KFS ranges from 0 to 1. The higher the criticality, the higher the score.

[0101] Historical Failure Rate Score (HFRS): Based on historical failure data, the failure rate of each component is calculated, and the failure rate score ranges from 0 to 1:

[0102]

[0103] Repair Cost Score (MCS): A score is given based on the cost of repairing or replacing parts. C represents repair cost. The repair cost score ranges from 0 to 1:

[0104] MCS=1-e -C ;

[0105] Usage Frequency Score (UFS): A component is scored based on how often it is used and the workload it is under. The Usage Frequency Score ranges from 0 to 1:

[0106]

[0107] Supply Chain Complexity Score (SCCS): Scored based on the complexity and stability of the supply chain, where n represents the number of suppliers and L represents the supplier level. The supply chain complexity score ranges from 0 to 1:

[0108]

[0109] User Feedback Rating (US): Rating is based on user feedback satisfaction and complaint rate. The user feedback rating ranges from 0 to 1:

[0110]

[0111] 3. Weight allocation. Invite 8 experts, including 2 experts in the fields of automotive engineering, parts manufacturing, automotive maintenance, and market analysis, to participate in determining the scoring weights for each dimension. First, conduct the first round of questionnaires and send questionnaires to the expert group to determine the initial scoring weights for each dimension. Collect the questionnaires and summarize them, calculate the average of the recommended scoring weights for each dimension, and feed back the calculation results to each expert. Start the second round of questionnaires and ask the experts to re-assign weights to each dimension based on the results of the previous round. Collect the questionnaires and count the results again, calculate the average and feed back the results to the experts. Continue to conduct three rounds of questionnaires and feedback the summary results between each round of surveys. Experts in each round can adjust their weight suggestions until the expert group reaches a certain consensus and the opinions tend to be stable. At this time, the final scoring weights for each dimension are output.

[0112] 4. Calculate the comprehensive score of each component based on the score and weight of each dimension.

[0113] 5. Based on the comprehensive score, the parts are divided into four grades: A, B, C, and D. Grade A: comprehensive score ≥ 0.8, Grade B: 0.8>comprehensive score ≥ 0.7, Grade C: 0.7>comprehensive score ≥ 0.5, Grade D: 0.5>comprehensive score.

[0114] In this embodiment, a machine learning algorithm is further used to dynamically adjust and optimize the scoring system, and the weight distribution of the scores of each dimension is optimized according to the continuous input of new data. The algorithm steps are as follows:

[0115] 1. Initially, the weights of the scoring parameters of each dimension have been determined as W 0 ={ω 0,1 ,ω 0,2 ,...ω 0,n}, where ω0,i Represents the initial proportion of the i-th parameter and satisfies:

[0116]

[0117] Scoring parameters include: key functionality score, historical failure rate score, repair cost score, usage frequency score, supply chain complexity score, and user feedback score;

[0118] 2. Collect and input new real-time data into the model. Each component will generate new real-time data during use. t ={d t,1 ,d t,2 ,...,d t,n}, where d t,i Represents the new data of the i-th parameter at time t. The new data is input into the scoring formula of each dimension to calculate the score, and the score of each dimension C is output t ={c t,1 ,c t,2 ,...,c t,n};

[0119] 3. Set a change sensitivity coefficient S for each scoring parameter i , which indicates the sensitivity of this parameter to the importance of the component, and the value range is 0 to 1. The higher the coefficient, the greater the impact of this parameter on the change of importance; the time weighting factor T is introduced t , whose value decreases over time, to reduce the impact of historical data on current calculations:

[0120]

[0121] Where λ is the decay rate parameter, t 0 is the initial time point;

[0122] 4. At each time point t, the weight of each dimension score is adjusted according to the deviation between the newly obtained score data of each dimension and the historical score data.

[0123] The adjustment formula is:

[0124]

[0125] Where: η is the learning rate, which controls the adjustment pace, is the historical average of the ith parameter, is the historical standard deviation of the i-th parameter; the adjusted score ratio is:

[0126] ω t+1,i =ω t,i +Δω t,i ;

[0127] 5. Normalize to ensure that the adjusted score ratio still satisfies the constraint of summing to 1:

[0128]

[0129] 6. Output the results.

[0130] What needs to be specifically explained in this embodiment is that, through the above-mentioned machine learning algorithm, it can be achieved that when the score of any component in a certain dimension continues to increase, the weight of the score will also continue to increase. Improving both aspects simultaneously can quickly improve the importance level of the component, so that the components that personnel believe to be unimportant based on experience are quickly taken seriously. When relevant personnel discover this situation, they should verify the situation of the component in a timely manner, and conduct troubleshooting and repair when necessary, and keep paying attention to it in the future; this method realizes intelligent sorting of the importance of automotive parts, and adjusts and updates them in real time. Through automated analysis, maintenance personnel can first check those components with higher importance when facing vehicle failures, so as to locate the problem faster and more accurately, and completely get rid of the traditional method of relying on human experience and judgment, and avoid misjudgment caused by lack of experience or subjective bias. This not only improves maintenance efficiency, but also significantly reduces the time and labor cost of troubleshooting, and provides a stronger guarantee for the safety and stability of the vehicle.

[0131] What needs to be further explained in this embodiment is that the importance of parts is determined by using a component importance evaluation and grading algorithm and divided into A, B, C, and D levels. The full life cycle information of the components of level A is generated into a hash value and stored on the blockchain network for subsequent comparison and verification to prevent information tampering. The blockchain type is a consortium chain, which is maintained by a dedicated person in the department. The information index and verification information on the consortium chain are associated with the component QR code and stored in the system database. After scanning the QR code with a scanning device, the component information can be retrieved through the consortium chain network. Using a machine learning algorithm, dynamic monitoring of components of level A is carried out, and the operation data of components is collected through sensors to establish a prediction model. According to the model budget results, important components are regularly inspected and reports are generated to prevent the vehicle from breaking down halfway due to sudden damage to components.

[0132] What needs to be further explained in this embodiment is that the workstation information, worker operation information, and the full life cycle information of parts with importance level A are generated and packaged in the block, and the blockchain type is a consortium chain; in order to reduce the data pressure on the chain, the workstation information and worker operation information are stored on one chain, and the full life cycle information of parts with importance level A is stored on another chain; each chain is maintained by a dedicated person from each department, and cross-chain data interaction is carried out by having a dedicated person from each department act as a notary, so that the information on the chain can flow across chains.

[0133] The vehicle fault inspection and analysis module analyzes the cause of the fault by collecting vehicle sensor information and fault codes (DTC). The module first monitors and summarizes the data from various vehicle sensors in real time. Through detailed analysis of these data, the module can identify potential problems and abnormal conditions. At the same time, fault codes (DTC, diagnostic fault codes) are also collected and interpreted to provide specific information about the fault. These fault codes serve as an important basis for diagnosis and help to accurately locate the source of the fault.

[0134] The information traceability module uses real-time diagnostic data and historical fault data, and uses the association rule mining Apriori algorithm to analyze the relationship between diagnostic codes and component failures. According to the importance level of the components and the specific fault conditions, high-importance and high-impact components are checked first; matching faults, using past fault information and common problems of the entire related vehicle model to establish a fault mode history library, matching current fault symptoms with historical fault modes, and quickly locating faulty components; using a scanning device to scan the code to obtain all relevant information of the component, including process design process information, processing process information, assembly process information, including transportation, installation and maintenance records, combined with the hash value on the blockchain to verify its complete life cycle information, and take corresponding processing measures.

[0135] In the real-time monitoring and feedback module, operators, inspectors, and on-site technicians have the authority to report any process or quality issues in the system. After the problem is reported, the task decomposition personnel will designate the responsible unit in the system to handle it. The responsible unit must provide feedback in the system after the handling is completed. The process can only be ended after the decomposition personnel confirms that the problem is closed. The system supports the decomposition of multiple responsible units for the same problem and data report query functions. During the operation of the vehicle, the status of key components is monitored in real time, and anomalies are discovered through sensor data. When the system detects an anomaly, it automatically generates an alarm to prompt the driver and maintenance personnel to check, and feedback the actual fault handling situation to the system to improve the fault mode history library and diagnostic model to prepare for the next precise fault positioning.

[0136] In this embodiment, Figure 2 A method for intelligent management and fault tracing of automobile parts includes the following steps:

[0137] S1. Data collection and storage: collect the production operation plan, supporting detailed data, and parts information of each vehicle, and integrate them into the single vehicle file storage based on the single vehicle VIN code; collect and store work station information, worker work information, and operation information;

[0138] S2. Component importance classification: dynamically rate the importance of automotive components, conduct regular inspections on A-level components, and generate reports;

[0139] S3, vehicle fault inspection and analysis, collecting vehicle sensor information and fault code DTC, and analyzing the cause of the fault;

[0140] S4. Information tracing, analyzing the relationship between fault codes and component failures, using a code scanning device to scan the code to obtain relevant information about the components, and generating maintenance suggestions and taking measures to deal with them;

[0141] S5. Real-time monitoring and feedback: real-time monitoring of the status of components during vehicle operation and generation of alarms. Real-time monitoring of the status of key components during vehicle operation. Operators have the authority to report any problems they find in the system. After the problems are reported, they will be assigned to designated responsible units for processing.

[0142] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent management and fault tracing system for automotive parts. It is characterized in that Includes the following modules: Data acquisition and storage module: used to collect and store information of various parts in the automobile production process; A component importance grading module uses a component importance grading algorithm to dynamically score and grade components based on multi-dimensional indicators, including component functions, failure history, usage frequency, maintenance cost, supply chain data, and user feedback. The grading is to divide components into different importance levels. Vehicle fault inspection and analysis module, used to perform fault analysis based on vehicle sensor information and fault codes; Information tracing module, used to troubleshoot component failures based on the failure mode history library and real-time diagnostic data, and generate maintenance recommendations; The real-time monitoring and feedback module is used to monitor the status of components during vehicle operation in real time and generate alarms.

2. The system according to claim 1, It is characterized in that The component importance grading algorithm comprises the following steps: 1) Collect functional information, historical failure information, maintenance cost information, usage frequency information, supply chain data information, and user feedback information of parts; 2) Establish a multi-dimensional scoring model to score each component based on its key functions, historical failure rate, usage frequency, supply chain complexity, maintenance cost, and user feedback; 3) Apply a weighted algorithm to the scoring results to calculate the importance score, and determine the importance level of the parts according to the preset threshold, and divide them into A, B, C, and D levels in descending order of importance; 4) Generate a hash value for the full life cycle information of grade A parts and store it on the blockchain network, and link it to the system database.

3. The system according to claim 1, It is characterized in that The data collection and storage module also collects the production operation plan and supporting detailed data of a single vehicle, integrates the single vehicle VIN code as the main body to form a single vehicle file, and stores the data in the system database and / or the cloud.

4. The system according to claim 1, It is characterized in that The data collection and storage module performs AES symmetrical encryption on the collected component information and generates a QR code, which is embedded on the surface of the component or on the packaging.

5. The system according to claim 1, It is characterized in that The information tracing module uses the fault mode history library to match the real-time diagnostic data to perform component fault troubleshooting and obtains component-related information through QR code scanning; the information tracing module uses an association rule mining algorithm to analyze the relationship between the fault code and the component failure to determine the source of the fault.

6. The system of claim 1, It is characterized in that The data collection and storage module collects and stores workstation information, worker operation information and component life cycle information, packages them for transactions through blockchain, and uses the alliance chain network for cross-chain data interaction; the alliance chain type is a specially maintained alliance chain, and the on-chain information index and verification information are associated with the component QR code to ensure the security and integrity of the data.

7. The system of claim 1, It is characterized in that In the component importance grading module, a machine learning algorithm is used to dynamically adjust and optimize the multi-dimensional scoring model, and the weight distribution of each dimension score is optimized according to the input of new data. The algorithm steps are as follows: S1. Initially, the weight of each dimension scoring parameter is determined as W 0 ={ω 0,1 ,ω 0,2 ,...ω 0,n }, where ω 0,i Represents the initial proportion of the i-th parameter and satisfies: S2, collect and input the new data generated by each component during use into the model in real time, D t ={d t,1 ,d t,2 ,...,d t,n }, where d t,i Represents the new data of the i-th parameter at time t. The new data is input into the scoring formula of each dimension to calculate the score, and the score of each dimension C is output t ={c t,1 ,c t,2 ,...,c t,n }; S3. Set a change sensitivity coefficient S for each scoring parameter i , the value range is 0 to 1. The higher the coefficient, the greater the impact of this parameter on the change of importance. The time weighting factor T is introduced. t , whose value decreases over time, to reduce the impact of historical data on current calculations: Where λ is the decay rate parameter, t 0 is the initial time point; S4. At each time point t, the weight of each dimension score is adjusted according to the deviation between the newly obtained score data of each dimension and the historical score data. The adjustment formula is: Where: η is the learning rate, which controls the adjustment pace, is the historical average of the ith parameter, is the historical standard deviation of the ith parameter; The adjusted score ratio is: oh t+1,i =ω t,i +See t,i ; S5. Normalize to ensure that the adjusted score ratio still satisfies the constraint that the sum is 1: S6: Result output.

8. The system of claim 7, It is characterized in that In the component importance grading module, the index and verification information on the alliance chain are associated with the component QR code and saved in the system database. After scanning the QR code, the full life cycle information of the component can be retrieved through the blockchain network. Using machine learning algorithms, dynamic monitoring of level A components is performed, component operation data is collected through sensors, and a prediction model is established. Important components are regularly inspected based on the model budget results, and inspection reports are generated.

9. The system according to claim 6, It is characterized in that The workstation information and worker operation information are stored on one blockchain, and the full life cycle information of parts with importance level A is stored on another blockchain; each blockchain is maintained by a dedicated person from each department, and cross-chain data interaction is carried out by having a dedicated person from each department act as a notary, allowing the information on the chain to flow across chains.

10. A method for intelligent management and fault tracing of automobile parts, the method being used to implement the system described in any one of claims 1 to 9 above, It is characterized in that The method comprises the following steps: Data collection and storage: collect the production operation plan, supporting detailed data, and parts information of each vehicle, and integrate them into single-vehicle file storage based on the single-vehicle VIN code; collect and store work station information, worker work information, and operation information; Component importance grading: dynamically grading the importance of automotive components, regularly inspecting A-level components, and generating reports; Vehicle fault inspection and analysis, collecting vehicle sensor information and fault codes, and analyzing fault causes; Information tracing, analyzing the relationship between fault codes and component failures, scanning codes to obtain relevant information about components, and generating maintenance recommendations; Real-time monitoring and feedback: real-time monitoring of the status of components during vehicle operation and generation of alarms.

Citation Information

Cited By

  • Motor vehicle maintenance management service system

    CN119692985A

  • Intelligent monitoring system and method for numerical control machine tool

    CN120190670A

  • An intelligent monitoring system and method for a numerically controlled machine tool

    CN120190670B

  • Property management intelligent scheduling and maintenance method and system

    CN120278437A

  • Quality data analysis management system based on gas appliance production and after-sales service

    CN121146623A